The principal goal of our randomized trial is to compare the efficacy of four different interventions, including conventional diet, conventional diet plus glipizide, conventional diet plus metformin, and a life style intervention with intensive diet and exercise, in preventing the progression of impaired glucose tolerance (IGT) to non-insulin dependent diabetes mellitus (NIDDM). Secondary goals include determining whether the interventions will affect the rate of reversion from IGT to normal glucose tolerance and/or improve cardiovascular risk factors, morbidity, or mortality. IGT prevails in 16% of the 40 to 74 year old U.S. population and is an appropriate target for intervention because IGT is l) a major risk factor for NIDDM which is difficult to treat effectively once it develops and is the cause of significant morbidity and mortality and 2), accompanied by numerous cardiovascular risk factors, morbidity and mortality. The proposal defines selection criteria for the IGT study population that will maximize the expected rate of conversion to NIDDM, empowering the study (75 to 93% power) to detect a 33 to 40% reduction in the hazard rate for development of NIDDM with intensive life-style intervention compared with the conventional diet. The selection criteria include screening for IGT subjects in populations with a high rate of recognized NIDDM (hispanic, black, and family members of NIDDM patients) and further selecting a subset of IGT subjects with a relatively high likelihood of progressing to NIDDM, i.e., obese IGT subjects with glucose levels equal to or greater than the median of the IGT range with OGTT. The experimental interventions have been selected based on 1) demonstrated salutary effects on glycemia and/or cardiovascular risk profile; 2) potential for practical application in the clinical setting;3) absence of frequent or severe side- effects. Investigators and subjects will be masked to treatment assignment except for the life-style intervention. Life-table analyses will be used to compare the cumulative incidences of the principal and secondary outcomes among the treatment groups. In addition, the absolute-level of chronic glycemia, selected lipid and insulin values, and blood pressure will be studied.